or it's being taught to the test / overfitting on the benches
November 14, 2026 at 11:56 PM
or it's being taught to the test / overfitting on the benches
Overfitting Random Fourier Features: Universal Approximation Property
https://thierrymoudiki.github.io/blog/2025/12/13/r/python/Overfitting-Random-Fourier-Features
#Python #DataScience #MachineLearning #rstats #Techtonique
https://thierrymoudiki.github.io/blog/2025/12/13/r/python/Overfitting-Random-Fourier-Features
#Python #DataScience #MachineLearning #rstats #Techtonique
November 15, 2026 at 3:08 PM
Overfitting Random Fourier Features: Universal Approximation Property
https://thierrymoudiki.github.io/blog/2025/12/13/r/python/Overfitting-Random-Fourier-Features
#Python #DataScience #MachineLearning #rstats #Techtonique
https://thierrymoudiki.github.io/blog/2025/12/13/r/python/Overfitting-Random-Fourier-Features
#Python #DataScience #MachineLearning #rstats #Techtonique
yeah, codependent overfitting. this is why ceos should be subject to dropout regularization and bagging
November 13, 2026 at 4:11 PM
yeah, codependent overfitting. this is why ceos should be subject to dropout regularization and bagging
Time Series Overfitting Gym quantlabs.co/horizon/
November 10, 2026 at 7:55 PM
Time Series Overfitting Gym quantlabs.co/horizon/
I mean, the public impression isn’t *wrong*, it just might be a bit of an overfitting of pattern in some cases (including this one). It’s still unsuitable for all applications (except domain-specific, human-curated ones which I think we’re better off naming as “ML”)
November 15, 2026 at 2:02 AM
I mean, the public impression isn’t *wrong*, it just might be a bit of an overfitting of pattern in some cases (including this one). It’s still unsuitable for all applications (except domain-specific, human-curated ones which I think we’re better off naming as “ML”)
overfitting is in every 101-level machine learning classes i've taken and this is always telling when anyone tells you to just "train them on each other bro"
November 12, 2026 at 12:58 PM
overfitting is in every 101-level machine learning classes i've taken and this is always telling when anyone tells you to just "train them on each other bro"
He's genuinely very good at applied information theory, has a good sense of when he'd be overfitting, what inherent uncertainty limits, just none of that means your opinions are good
November 10, 2026 at 4:22 PM
He's genuinely very good at applied information theory, has a good sense of when he'd be overfitting, what inherent uncertainty limits, just none of that means your opinions are good
Giuseppe D'Ambruoso, Giovanni Gramegna, Domenico Pomarico, Giuseppe Magnifico, Cosmo Lupo
Invariance can prevent overfitting and shortcut learning in Quantum Neural Networks
https://arxiv.org/abs/2611.14368
Invariance can prevent overfitting and shortcut learning in Quantum Neural Networks
https://arxiv.org/abs/2611.14368
November 12, 2026 at 5:35 AM
Giuseppe D'Ambruoso, Giovanni Gramegna, Domenico Pomarico, Giuseppe Magnifico, Cosmo Lupo
Invariance can prevent overfitting and shortcut learning in Quantum Neural Networks
https://arxiv.org/abs/2611.14368
Invariance can prevent overfitting and shortcut learning in Quantum Neural Networks
https://arxiv.org/abs/2611.14368
Next-Level Benchmarking Overfitting – Thomas Neumann | Discussion
Next-Level Benchmarking Overfitting
A blog by and for database architects.
databasearchitects.blogspot.com
November 11, 2026 at 1:40 PM
Next-Level Benchmarking Overfitting – Thomas Neumann | Discussion
Next-Level Benchmarking Overfitting – Thomas Neumann
http://databasearchitects.blogspot.com/2026/11/next-level-benchmarking-overfitting.html
[comments] [22 points]
http://databasearchitects.blogspot.com/2026/11/next-level-benchmarking-overfitting.html
[comments] [22 points]
Database Architects: Next-Level Benchmarking Overfitting
databasearchitects.blogspot.com
November 11, 2026 at 2:03 PM
Next-Level Benchmarking Overfitting – Thomas Neumann
http://databasearchitects.blogspot.com/2026/11/next-level-benchmarking-overfitting.html
[comments] [22 points]
http://databasearchitects.blogspot.com/2026/11/next-level-benchmarking-overfitting.html
[comments] [22 points]
My own forecast: once they’re heavily integrated into the system their forecasting is probably gonna turn to custard due to overfitting and model collapse.
Once you’re a part of the system you’re trying model all bets are off.
Once you’re a part of the system you’re trying model all bets are off.
AIs are making it easier for everyone to predict the future. That leaves open how far forecasters—human and machine—are from the theoretical limit of what can be known about the future
Artificial intelligence now beats some of the best human forecasters
Crystal balls give way to LLMs
bit.ly
November 10, 2026 at 6:02 AM
My own forecast: once they’re heavily integrated into the system their forecasting is probably gonna turn to custard due to overfitting and model collapse.
Once you’re a part of the system you’re trying model all bets are off.
Once you’re a part of the system you’re trying model all bets are off.
Next-Level Benchmarking Overfitting – Thomas Neumann http://databasearchitects.blogspot.com/2026/11/next-level-benchmarking-overfitting.html (https://news.ycombinator.com/item?id=50452496)
Next-Level Benchmarking Overfitting
A blog by and for database architects.
databasearchitects.blogspot.com
November 11, 2026 at 5:17 PM
Next-Level Benchmarking Overfitting – Thomas Neumann http://databasearchitects.blogspot.com/2026/11/next-level-benchmarking-overfitting.html (https://news.ycombinator.com/item?id=50452496)
Next-Level Benchmarking Overfitting
A blog by and for database architects.
databasearchitects.blogspot.com
November 10, 2026 at 12:25 PM
OF COURSE THEY DO THIS IS CALLED OVERFITTING AND IT IS A KNOWN FOREVER PROBLEM WITH MACHINE LEARNING
November 10, 2026 at 3:29 AM
OF COURSE THEY DO THIS IS CALLED OVERFITTING AND IT IS A KNOWN FOREVER PROBLEM WITH MACHINE LEARNING
💡 Overfitting: the model is fitting noise in the training data rather than patterns that generalize. Keep the weights from the best validation epoch (early stopping), and try regularization or more data.
How would you tell real overfitting from a noisy bump?
How would you tell real overfitting from a noisy bump?
November 12, 2026 at 2:00 PM
💡 Overfitting: the model is fitting noise in the training data rather than patterns that generalize. Keep the weights from the best validation epoch (early stopping), and try regularization or more data.
How would you tell real overfitting from a noisy bump?
How would you tell real overfitting from a noisy bump?
Next-Level Benchmarking Overfitting
Comments
databasearchitects.blogspot.com
November 10, 2026 at 4:07 PM
Next-Level Benchmarking Overfitting
A blog by and for database architects.
databasearchitects.blogspot.com
November 10, 2026 at 12:37 PM
You'll also learn how to evaluate model performance, avoid common pitfalls such as overfitting, and assess whether your models can make reliable predictions on new data.
November 13, 2026 at 1:42 PM
You'll also learn how to evaluate model performance, avoid common pitfalls such as overfitting, and assess whether your models can make reliable predictions on new data.
Backtesting Your Crypto Bot Without Fooling Yourself: The 2026 Practitioner's Guide
Backtesting Your Crypto Bot Without Fooling Yourself: The 2026 Practitioner's Guide
A passing backtest doesn't mean a profitable bot. Overfitting, lookahead bias, and fee blindness kill more strategies than bad signals. Here's how to build a backtest that actually means something in 2026.
donperro.io
November 14, 2026 at 8:14 PM
Backtesting Your Crypto Bot Without Fooling Yourself: The 2026 Practitioner's Guide
Next-Level Benchmarking Overfitting – Thomas Neumann
http://databasearchitects.blogspot.com/2026/11/next-level-benchmarking-overfitting.html
http://databasearchitects.blogspot.com/2026/11/next-level-benchmarking-overfitting.html
November 11, 2026 at 1:10 PM
Next-Level Benchmarking Overfitting – Thomas Neumann
http://databasearchitects.blogspot.com/2026/11/next-level-benchmarking-overfitting.html
http://databasearchitects.blogspot.com/2026/11/next-level-benchmarking-overfitting.html
Next-Level Benchmarking Overfitting – Thomas Neumann
#HackerNews
<a href="http://databasearchitects.blogspot.com/2026/11/next-level-benchmarking-overfitting.html" class="hover:underline text-blue-600 dark:text-sky-400 no-card-link" target="_blank" rel="noopener" data-link="bsky">http://databasearchitects.blogspot.com/2026/11/next-level-benchmarking-overfitting.html
#HackerNews
<a href="http://databasearchitects.blogspot.com/2026/11/next-level-benchmarking-overfitting.html" class="hover:underline text-blue-600 dark:text-sky-400 no-card-link" target="_blank" rel="noopener" data-link="bsky">http://databasearchitects.blogspot.com/2026/11/next-level-benchmarking-overfitting.html
November 11, 2026 at 2:47 PM
Next-Level Benchmarking Overfitting – Thomas Neumann
#HackerNews
<a href="http://databasearchitects.blogspot.com/2026/11/next-level-benchmarking-overfitting.html" class="hover:underline text-blue-600 dark:text-sky-400 no-card-link" target="_blank" rel="noopener" data-link="bsky">http://databasearchitects.blogspot.com/2026/11/next-level-benchmarking-overfitting.html
#HackerNews
<a href="http://databasearchitects.blogspot.com/2026/11/next-level-benchmarking-overfitting.html" class="hover:underline text-blue-600 dark:text-sky-400 no-card-link" target="_blank" rel="noopener" data-link="bsky">http://databasearchitects.blogspot.com/2026/11/next-level-benchmarking-overfitting.html
Next-Level Benchmarking Overfitting – Thomas Neumann
comments · posted on 2026.11.11 at 07:09:49 (c=0, p=3)
comments · posted on 2026.11.11 at 07:09:49 (c=0, p=3)
November 11, 2026 at 12:59 PM
Next-Level Benchmarking Overfitting – Thomas Neumann
comments · posted on 2026.11.11 at 07:09:49 (c=0, p=3)
comments · posted on 2026.11.11 at 07:09:49 (c=0, p=3)
Worth a read: Thomas Neumann writes about next-level benchmarking overfitting.
Also: Ocient raised a $100M Series C in Chicago. And Layup is hiring a Software Engineer (Computational Geometry) in Huntington Beach.
theconsensus.dev/g/2026/11/09/
Also: Ocient raised a $100M Series C in Chicago. And Layup is hiring a Software Engineer (Computational Geometry) in Huntington Beach.
theconsensus.dev/g/2026/11/09/
November 9, 2026 at 4:51 PM
Worth a read: Thomas Neumann writes about next-level benchmarking overfitting.
Also: Ocient raised a $100M Series C in Chicago. And Layup is hiring a Software Engineer (Computational Geometry) in Huntington Beach.
theconsensus.dev/g/2026/11/09/
Also: Ocient raised a $100M Series C in Chicago. And Layup is hiring a Software Engineer (Computational Geometry) in Huntington Beach.
theconsensus.dev/g/2026/11/09/
Absolutely. The real challenge is detecting regime transitions early enough while avoiding false switches and overfitting. Probabilistic regime detection, adaptive thresholds and transaction costs are key. Which indicators or models have you found most reliable?
November 14, 2026 at 11:25 PM
Absolutely. The real challenge is detecting regime transitions early enough while avoiding false switches and overfitting. Probabilistic regime detection, adaptive thresholds and transaction costs are key. Which indicators or models have you found most reliable?